Case studies · free

Three content operations, taken apart.

Same five steps each time, ending with what the correct plan does to the numbers. Where a figure is ours we say so and show the arithmetic.

Any B2B publisherongoing
Production

Twenty pieces a month, and nothing compounds

  1. What was done

    A team commits to twenty posts a month. Topics come from a keyword export, each writer takes a row, and the piece goes out when it is finished.

  2. What was right

    The publishing discipline itself is real. Hitting a cadence for a year is harder than it looks, and the team built a working production line.

  3. What went wrong

    No brief, so no piece answers a question anybody named. No original figure, so every piece is a summary of something the reader could have read instead. Nothing is reviewed after publishing, so the archive decays while the team adds to it.

  4. What should have happened

    Cut to eight pieces a month. Each gets a one page brief with the question, the audience and one figure the piece must produce. Add a review date at publication and an inventory of what already exists.

What the right plan does to the economics

Our own arithmetic on a modelled team: 20 pieces a month at roughly 600 dollars all-in each is 144,000 dollars a year. Cutting to 8 pieces frees 86,400 dollars, and spending a third of that on research and fact checking still leaves about 57,000. The output that gets cited is the third with original figures in it, which under the old plan was none of them.

Our own modelled team and arithmetic, offered as a worked example rather than a case
Any content siteongoing
Structure

Four pages on one question, and none of them rank

  1. What was done

    Over three years the team publishes four separate pieces that all half answer the same question, each written by somebody who did not know the others existed.

  2. What was right

    All four are competently written and each has something the others do not. Nothing here is a quality problem.

  3. What went wrong

    The search engine picks one and usually the weakest, because links and internal signals split four ways. Traffic does not divide evenly between them, it collapses towards the survivor.

  4. What should have happened

    Search your own domain for the term, list every page that appears, pick the survivor by links pointing at it rather than by preference, move the useful parts across and redirect the rest.

What the right plan does to the economics

On our own five site network we measured the same class of problem: an average page had 8.2 inbound internal links but only 2.6 came from inside body text, the rest were header and footer. Two of fifteen pages had zero contextual inbound links at all. A merge plus two contextual links is an afternoon of work and costs nothing, against the 2,400 dollars four pieces cost to write.

Our own measurement across five sites using tools/linkgraph.py · figures are ours, method described in the piece
Any team under pressureongoing
Maintenance

The refresh that changed only the date

  1. What was done

    Traffic falls, so the team runs a refresh programme: fifty old posts get a new publication date and two reordered paragraphs each.

  2. What was right

    Recognising that the archive needed attention was correct. Most teams keep publishing new pieces instead, which is the more expensive mistake.

  3. What went wrong

    Nothing new was added, so nothing changed for the reader, and readers can tell within one screen. Search engines see the same content with a new date, which is a signal they have handled for years.

  4. What should have happened

    Refresh means one new figure, removal of advice that stopped being true, and a line saying what changed and when. Fifteen genuinely refreshed pieces beat fifty redated ones.

What the right plan does to the economics

Our own arithmetic: fifty redated posts at two hours each is 100 hours. At a 60 dollar hourly rate that is 6,000 dollars spent to change fifty dates. The same 100 hours across fifteen real refreshes is under seven hours each, enough to find a new number and rewrite the weakest section, which is the version that holds position.

Our own arithmetic on a modelled team · Google Search Central guidance on content freshness

No figure here is attributed to a named company. Everything is our own arithmetic on modelled teams, or our own measurement across our own sites, and each case says which. Teaching material, not advice for your business.

Free consultation

Recognise your team in one of these?

Tell us which one and how much you publish. An editor replies with what to change first. No call.

Build a brief free